DEV Community

Super Kai (Kazuya Ito)
Super Kai (Kazuya Ito)

Posted on • Edited on

sub and mul in PyTorch

#python #pytorch #subtraction #multiplication

Buy Me a Coffee☕

*Memos:

sub() can do subtraction with two of the 0D or more D tensors of zero or more elements or scalars or the 0D or more D tensor of zero or more elements and a scalar, getting the 0D or more D tensor of zero or more elements as shown below:

*Memos:

  • sub() can be used with torch or a tensor.
  • The 1st argument(input) with torch(Type:tensor or scalar of int, float or complex) or using a tensor(Type:tensor of int, float or complex)(Required).
  • The 2nd argument with torch or the 1st argument with a tensor is other(Required-Type:tensor or scalar of int, float or complex).
  • The 3rd argument with torch or the 2nd argument with a tensor is alpha(Optional-Default:1-Type:tensor or scalar of int, float or complex). *otheris multiplied by alpha(input or a tensor-(otherxalpha)).
  • There is out argument with torch(Optional-Default:None-Type:tensor): *Memos:
    • out= must be used.
    • My post explains out argument.
  • subtract() is the alias of sub(). 
import torch

tensor1 = torch.tensor([9, 7, 6])
tensor2 = torch.tensor([[4, -4, 3], [-2, 5, -5]])

torch.sub(input=tensor1, other=tensor2)
tensor1.sub(other=tensor2)
torch.sub(input=tensor1, other=tensor2, alpha=1)
torch.sub(input=tensor1, other=tensor2, alpha=torch.tensor(1))
# tensor([[5, 11, 3], [11, 2, 11]])

torch.sub(input=tensor1, other=tensor2, alpha=0)
torch.sub(input=tensor1, other=tensor2, alpha=torch.tensor(0))
# tensor([[9, 7, 6], [9, 7, 6]])

torch.sub(input=tensor1, other=tensor2, alpha=2)
torch.sub(input=tensor1, other=tensor2, alpha=torch.tensor(2))
# tensor([[1, 15, 0], [13, -3, 16]])

torch.sub(input=tensor1, other=tensor2, alpha=-1)
torch.sub(input=tensor1, other=tensor2, alpha=torch.tensor(-1))
# tensor([[13, 3, 9], [7, 12, 1]])

torch.sub(input=tensor1, other=tensor2, alpha=-2)
torch.sub(input=tensor1, other=tensor2, alpha=torch.tensor(-2))
# tensor([[17, -1, 12], [5, 17, -4]])

torch.sub(input=9, other=tensor2)
torch.sub(input=9, other=tensor2, alpha=1)
torch.sub(input=9, other=tensor2, alpha=torch.tensor(1))
# tensor([[5, 13, 6], [11, 4, 14]])

torch.sub(input=tensor1, other=4)
torch.sub(input=tensor1, other=4, alpha=1)
torch.sub(input=tensor1, other=4, alpha=torch.tensor(1))
# tensor([5, 3, 2])

torch.sub(input=9, other=4)
torch.sub(input=9, other=4, alpha=1)
torch.sub(input=9, other=4, alpha=torch.tensor(1))
# tensor(5)

tensor1 = torch.tensor([9., 7., 6.])
tensor2 = torch.tensor([[4., -4., 3.], [-2., 5., -5.]])

torch.sub(input=tensor1, other=tensor2)
torch.sub(input=tensor1, other=tensor2, alpha=1.)
torch.sub(input=tensor1, other=tensor2, alpha=torch.tensor(1.))
# tensor([[5., 11., 3.], [11., 2., 11.]])

torch.sub(input=9., other=tensor2)
torch.sub(input=9., other=tensor2, alpha=1.)
torch.sub(input=9., other=tensor2, alpha=torch.tensor(1.))
# tensor([[5., 13., 6.], [11., 4., 14.]])

torch.sub(input=tensor1, other=4)
torch.sub(input=tensor1, other=4, alpha=1.)
torch.sub(input=tensor1, other=4, alpha=torch.tensor(1.))
# tensor([5., 3., 2.])

torch.sub(input=9., other=4)
torch.sub(input=9., other=4, alpha=1.)
torch.sub(input=9., other=4, alpha=torch.tensor(1.))
# tensor(5.)

tensor1 = torch.tensor([9.+0.j, 7.+0.j, 6.+0.j])
tensor2 = torch.tensor([[4.+0.j, -4.+0.j, 3.+0.j],
                        [-2.+0.j, 5.+0.j, -5.+0.j]])
torch.sub(input=tensor1, other=tensor2)
torch.sub(input=tensor1, other=tensor2, alpha=1.+0.j)
torch.sub(input=tensor1, other=tensor2, alpha=torch.tensor(1.+0.j))
# tensor([[5.+0.j, 11.+0.j, 3.+0.j],
#         [11.+0.j, 2.+0.j, 11.+0.j]])

torch.sub(input=9.+0.j, other=tensor2)
torch.sub(input=9.+0.j, other=tensor2, alpha=1.+0.j)
torch.sub(input=9.+0.j, other=tensor2, alpha=torch.tensor(1.+0.j))
# tensor([[5.+0.j, 13.+0.j, 6.+0.j],
#         [11.+0.j, 4.+0.j, 14.+0.j]])

torch.sub(input=tensor1, other=4.+0.j)
torch.sub(input=tensor1, other=4.+0.j, alpha=1.+0.j)
torch.sub(input=tensor1, other=4.+0.j, alpha=torch.tensor(1.+0.j))
# tensor([5.+0.j, 3.+0.j, 2.+0.j])

torch.sub(input=9.+0.j, other=4.+0.j)
torch.sub(input=9.+0.j, other=4.+0.j, alpha=1.+0.j)
torch.sub(input=9.+0.j, other=4.+0.j, alpha=torch.tensor(1.+0.j))
# tensor(5.+0.j)
Enter fullscreen mode Exit fullscreen mode

mul() can do multiplication with two of the 0D or more D tensors of zero or more elements or scalars or the 0D or more D tensor of zero or more elements and a scalar. getting the 0D or more D tensor of zero or more elements as shown below:

*Memos:

  • mul() can be used with torch or a tensor.
  • The 1st argument(input) with torch(Type:tensor or scalar of int, float, complex or bool) or using a tensor(Type:tensor of int, float, complex or bool)(Required).
  • The 2nd argument with torch or the 1st argument with a tensor is other(Required-Type:tensor or scalar of int, float, complex or bool).
  • There is out argument with torch(Optional-Default:None-Type:tensor): *Memos:
    • out= must be used.
    • My post explains out argument.
  • multiply() is the alias of mul(). 
import torch

tensor1 = torch.tensor([9, 7, 6])
tensor2 = torch.tensor([[4, -4, 3], [-2, 5, -5]])

torch.mul(input=tensor1, other=tensor2)
tensor1.mul(other=tensor2)
# tensor([[36, -28, 18], [-18, 35, -30]])

torch.mul(input=9, other=tensor2)
# tensor([[36, -36, 27], [-18, 45, -45]])

torch.mul(input=tensor1, other=4)
# tensor([36, 28, 24])

torch.mul(input=9, other=4)
# tensor(36)

tensor1 = torch.tensor([9., 7., 6.])
tensor2 = torch.tensor([[4., -4., 3.], [-2., 5., -5.]])

torch.mul(input=tensor1, other=tensor2)
# tensor([[36., -28., 18.], [-18., 35., -30.]])

torch.mul(input=9., other=tensor2)
# tensor([[36., -36., 27.], [-18., 45., -45.]])

torch.mul(input=tensor1, other=4.)
# tensor([36., 28., 24.])

torch.mul(input=9., other=4.)
# tensor(36.)

tensor1 = torch.tensor([9.+0.j, 7.+0.j, 6.+0.j])
tensor2 = torch.tensor([[4.+0.j, -4.+0.j, 3.+0.j],
                        [-2.+0.j, 5.+0.j, -5.+0.j]])
torch.mul(input=tensor1, other=tensor2)
# tensor([[36.+0.j, -28.+0.j, 18.+0.j],
#         [-18.+0.j, 35.+0.j, -30.+0.j]])

torch.mul(input=9.+0.j, other=tensor2)
# tensor([[36.+0.j, -36.+0.j, 27.+0.j],
#         [-18.+0.j, 45.+0.j, -45.+0.j]])

torch.mul(input=tensor1, other=4.+0.j)
# tensor([36.+0.j, 28.+0.j, 24.+0.j])

torch.mul(input=9.+0.j, other=4.+0.j)
# tensor(36.+0.j)

tensor1 = torch.tensor([True, False, True])
tensor2 = torch.tensor([[False, True, False], [True, False, True]])

torch.mul(input=tensor1, other=tensor2)
# tensor([[False, False, False],
#         [True, False, True]])

torch.mul(input=True, other=tensor2)
# tensor([[False, True, False], [True, False, True]])

torch.mul(input=tensor1, other=False)
# tensor([False, False, False])

torch.mul(input=True, other=False)
# tensor(False)
Enter fullscreen mode Exit fullscreen mode

Top comments (0)

Read next

grantdotdev profile image

Advent of Code 2024 - Day 3: Mull it Over

Grant Riordan -

fast profile image

Guide to 24 Essential Open Source Projects from Package Managers to AI apps

fast-d3v -

perisicnikola37 profile image

Automated crypto price tracking using GMAIL and Python

Nikola Perišić -

ppaanngggg profile image

YOLOv11: A New Breakthrough in Document Layout Analysis

ppaanngggg -